What David Roberts Gets Wrong about “Buying” Green Power

November 12, 2015, by Michael Gillenwater

Vox columnist David Roberts, who has spent years as an environmental blogger digging in on abstruse climate and energy issues and is one of the best of his kind on those subjects, recently dove into the important and complex issue of green power purchasing and Renewable Energy Certificates, or RECs. “RECs, which put the “green” in green electricity, explained” (9 November 2015).

Roberts asks the question: How can consumers buy green power? He writes to explain RECs and green power, but it ultimately reaches an erroneous and damaging conclusion, arguing that buying so-called “voluntary,” or unregulated, RECs has a meaningful benefit in the fight against climate change. This is incorrect, and disappointingly, Roberts fails to offer evidence or a logical argument in support his conclusion. My colleagues and I have studied RECs and green power for years. We can objectively say without hesitation that the voluntary REC market, with rare exceptions, has no effect on renewable energy generation or investment in the United States, nor is there evidence to indicate it does outside of the United States. Below, I’ll explain why, and provide confirming evidence.

A voluntary REC is a claim on the green attributes of electricity that isn’t meaningfully subject to government regulation. (Regulated RECs are also used to help meet state renewable energy standards. These are a different beast, are much more expensive, and can make a difference when government mandates force additional renewable energy generation. Unfortunately, Roberts article confuses his readers on this important distinction.) The dominant practice is for companies and families that wish to make a green power purchasing claim to use voluntary RECs, whether or not they realize it. Despite their stated definition and endorsement by well-intentioned organizations, RECs do not credibly represent the purchase of green power because they have no influence on renewable energy investment or generation. Roberts starts by making this case as well, but ultimately (and somewhat oddly) reaches the opposite conclusion, that we should voluntarily buy RECs. And he does so without evidence to support his argument. I want to carefully unpack how he finds his way to this conclusion because there is much we can learn from this example. And if we waste resources on an ineffective mechanism, we can’t spend it on solutions that lead to actual environmental benefits.

Roberts rightly points out that end-users literally cannot buy green power, any more than you can dip a cup into a lake and draw out the Perrier someone poured in at the other end. So he is brilliantly correct there. The physics of the grid should make this obvious, but it is rarely recognized. (Although, Roberts would benefit from some tutoring on electrons and the physics of electricity distribution.)

Next, Roberts argues that RECs are how we work-around the physics problem, and he cites NREL and US EPA green power marketing materials for support. This appeal to authority fallacy is the norm with green power marketing. I doubt that Roberts would be so easily fooled by the USDA pork or dairy promotion board by claims of the miracle health benefits of bacon and butter. He ought to know that you can’t always trust public benefit claims by government agencies that have as their mission the promotion of a particular industry.

Never mind all that. What Roberts misses is the proper question: Are voluntary RECs a legitimate work-around to the physics of the grid. Or better put: Does the voluntary REC market effectively simulate a consumer market for green electricity?

To answer this question we have to understand what a REC is, which Roberts attempts to explain. He tells us that a REC is defined as representing the social and environmental benefits or attributes of electricity produced with renewable energy. Okay, so what are these benefits or attributes and what exactly does it mean to commodify, trade, and buy them? Are attributes equal to green power?[1] Roberts ignores these questions, but then posits: “An REC is effectively a certificate of property rights over one unit of greenness.” In reality, what exists is unregulated market that sells “greenness” certificates, without a sound economic or legal basis for what greenness property rights means in theory or practice.[2] This point is where the legal or ethical concerns arise, in the same way that health claims about nutritional supplements are so sketchy.

Next, Roberts repeats the conventional magical thinking on RECs in that when traded and sold they turn whoever they land with into pollution free energy consumer. And then his article dives into an intellectual swamp:

Now, there’s nothing intrinsically shady about any of this. It is simply a way of giving the positive externalities of renewable energy generation a tangible representation in the marketplace. Where questions about RECs arise is in the way they are used and the claims that some people make on their behalf.

The truth is that the evidence demonstrates that there is something shady going on within the voluntary RECs industry. Yes, part of the problem is the deceptive claims made by those marketing and using them to crow in sustainability reports about their lowered corporate carbon footprints. But, there is also an intrinsic problem with voluntary RECs. Externalities don’t magically attach themselves to a certificate because you want them to. An externality is, by definition, a public good or harm that is not captured by the market. Simply because I start printing virtual certificates does not suddenly cause the value of these externalities to be embedded in the certificates. That is green magical thinking, not economics.

Environmental commodities in a voluntary market context (i.e., where there isn’t an enforced cap or quota that the commodity is being used to meet) can only function as a subsidy to the physical goods or services we agree are environmentally superior. This what a voluntary REC is supposed to do. But the subsidy then needs to actually alter the market for there to be any avoided externality. In the case of green power, if there is no change in the market for renewable electricity generation due to voluntary REC purchases, then there is no avoided externality, and then there is no “benefit” for voluntary RECs to represent. They are effectively nothing. Just wishful thinking and marketing spin, or to state formally, an ineffective charitable contribution.

So, to the crux question: does the voluntary REC market create a subsidy that does alter the renewable energy market, even just a little? Luckily, we have an empirical answer to this question based on peer-reviewed research. You can see the it here and here, but it can also be stated concisely: No! Based on market data and academic research by myself and others, the voluntary REC market (with rare exceptions) has no effect on renewable energy generation or investment.

Oddly, Roberts appears to be aware of the fact that the voluntary REC market is economically irrelevant. He even presents graphical data on the low prices (although these prices need to be put in a risk-adjusted context of overall renewable generation financing). He blames the low prices on lower standards for voluntary RECs compared to standards for compliance RECs. But it’s not the relative standards for compliance versus voluntary REC registries that are the meaningful difference here. It is the government mandate that forces utilities to buy compliance RECs from generators within certain states thereby creating a shortage, compared with nothing forcing electricity consumers to buy voluntary RECs. The fact is that lots of renewable energy projects have historically built without selling voluntary RECs, because there are many times more voluntary RECs available for sale nationally than there is demand from buyers. In such a market, it should not be surprising that prices are low. Why not zero? The economics answer is there are still transaction costs. (Someone has to pay for all that marketing.) Roberts seems to base his conclusion on the assumption that since RECs are worth something, they must at least have some small influence on renewable energy investment. Yet, once we consider transaction costs and renewable energy project finance, this assumption does not hold up when tested against the empirical evidence.

Next, Roberts correctly points out that a voluntary REC does not guarantee any particular amount of avoided emissions or other environmental benefits. (So what do they represent? Doesn’t that contradict his definition?) He also points out that voluntary RECs are not carbon offsets (Hallelujah that we have finally won that battle with Green-e and other green power marketers who claimed otherwise for years!!!) But then, his article paradoxically assumes that each MWh associated with a voluntary REC is avoiding an MWh of fossil fuel-fired generation. This is painful. Roberts recognizes that RECs are not offsets and that they don’t guarantee that any particular amount of dirty electricity generation is avoided. And then next paragraph supports the “approximate” claim that each REC prevents the generation of an MWh of electricity from dirty generation sources. I’m missing the logic there.

Yes, the growth of renewable energy at the expense of polluting generation sources is a good thing. And more renewable energy replacing fossil energy means less pollution. But, I think we knew this fact already. The relevant question is instead whether the voluntary REC market influences the amount of renewable energy generation. Again, we know the answer to this question empirically: “no.” And the academic debate is simplified because there is no research evidence to the contrary. I would cite the competing studies and engage in a proper academic debate on research methodology, but there are no studies showing otherwise to consider. Ask those promoting voluntary REC markets and you will go in circles with more logical fallacies, but be provided no evidence. Indeed, one only need get on the phone with REC middleman[3] and ask for a pro forma showing how voluntary RECs influences clean energy project finances. Hint: no substantiating information will be forthcoming.

The lowering cost of wind turbines is a wonderful development and it is not the cause of the problem with the voluntary REC market. The problem is with the thinking that voluntary RECs create a legitimate simulation of a market that allows consumers to buy green power. Indeed, Roberts concedes that voluntary RECs create no additionality in renewable energy markets! He says that “everyone should be clear what they’re getting.” Horray! But, then comes a backflip landing on his article’s mental gymnastics. He coins the term “indirect additionality,”[4] claiming that while voluntary RECs do not influence individual renewable energy investments that somehow they do still indirectly influence it on a “total market basis.” Okay? How so? Where is the data to back that assertion? Research shows that commodity voluntary RECs both at the individual project level and the overall market level, do not influence renewable energy investment or generation. Further, modeling shows that is no practical scenario where future grown in the demand for voluntary RECs will increase prices enough to change this conclusion (see here). Also see Is your “green power” really just “green washing?”And Have you fallen for the green power accounting shell game?

Roberts next admirably questions corporate green power purchasing claims and recognizes the serious potential for green washing. But then his article does an encore backflip by saying that companies should still go ahead and buy voluntary RECs solely because they are “cheap.” Supposedly this will all be clarified in his next post.

As for his proposed reforms, tweaking Green-e certificate requirements is not going to fix the fundamental problems. One could create a new type of certificate that was a truly additional REC (which has been done by a couple of firms under the label REC-plus and Gold Power), but that would be far more market tightening than anything Green-e would support. And this approach would effectively create a carbon offset credit denoted in MWh, which economically would appear silly because we already have something called a carbon offset credit (see here for a more detailed explanation).

Roberts’ second recommendation is a solution to the voluntary green power purchasing issue, because it would eliminate it. The recommendation would replace the voluntary market with an expanded (ideally federal) RPS and/or cap and trade policy. A public policy solution to a public goods problem.

But then if voluntary RECs are not a legitimate instrument for buying green power, then what is? This is an excellent question, especially at a time of experimentation with many different ways of structuring power purchasing agreements (PPAs) and green energy investments. Unfortunately, the answer to this question is not obvious. It is the subject of research we and others are currently engaged in. One thing is clear, though. Simplistically treating all these different financial instruments as “buying green power,” and thereby equivalent from an environmental accounting viewpoint, is very wrong. The goal of our research is to identify existing or new green power market instruments that represent a credible environmental benefit claim and formulate credible environmental accounting approaches so we can focus on voluntary actions that make a difference.

I’ll end my critique by summarizing the most important point: voluntary RECs do not have a marginal influence on the climate problem. The evidence confirms this conclusion. Current practice, blessed by well-intentioned government and NGO marketing, encourages companies to buy voluntary RECs and claim they are “100% green powered” and then attest to an erased carbon footprint. Effectively, they are then attesting to having reduced their carbon footprint. Problem solved, at least from a public relations standpoint. Yet, emissions to the atmosphere have not been reduced anywhere. Unfortunately, companies are being led into a devilishly bewildering greenwashing trap. Well-intentioned individuals think they’re doing good without spending much money, while meaningful actions that actually reduce carbon footprints are likely displaced.

I am passionate about this issue because I care deeply about the integrity of environmental accounting. The way countries, companies, projects and individuals measure and publicly attest to their environmental impacts needs to be meaningful. If we are really going to solve the climate problem, then we must have reasonable confidence in how we measure and track our performance. Currently, green power claims are an intellectually fascinating, but frightening case of deceptive environmental accounting that I fear will set a dangerous precedent for the future (see here for a deeper analysis).


I realize that my tone in this post has been somewhat antagonistic. I’ll admit to venting some frustration. Indeed, I continue to be aghast at what people assert about the validity of voluntary REC markets without being sufficiently skeptical. This said, my thoughtful colleague Derik Broekhoff (Stockholm Environment Institute) and I would welcome a chance to have a genuine discussion about these issues with Mr. Roberts. And it so happens that we all live in beautiful Seattle.

[1] If you want a deep dive on RECs and “attributes”, then I suggest you give this article a read: “Redefining RECs—Part 1: Untangling attributes and offsets

[2] What legal basis does exist draws from legislative and regulatory language intended for RPS compliance tracking.

[3] See Chapter 7 of Auden Schendler’s book Getting Green Done for a transcript of such a conversation.

[4] If you want a deep dive on the wacky and mentally sloppy ways that others try to describe additionality, and an explanation of how to understand this key environmental markets concept, see this article: “What is Additionality?: Part 1: A long standing problem.”

9 responses to “What David Roberts Gets Wrong about “Buying” Green Power”

  1. What about the accommodation for voluntary green power purchases in the Regional Greenhouse Gas Initiative (RGGI)? This part of the program was especially created to please the green power marketers, who because of it are able to claim that voluntary green power purchases actually do reduce GHGs. Under RGGI’s rules, there is a “carveout” from the program’s overall emission allowance budget – regulators intentionally set aside an amount of GHG allowances representing the emissions avoided by the amount of kilowatt hours of electricity voluntarily purchased within the nine RGGI states over a certain time period. After that time period, regulators look to see if that much renewable power was actually purchased and if it was, the set-aside amount is cancelled. Those allowances do not become available to emitters, thus the market is “short” that amount of supply, thus the price of emission allowances goes up, thus power generators pay more per ton GHG emitted. In this case, the voluntarily purchased power is actually reducing the regional GHG cap (and therefore actually incentivising clean power) so claims to that effect are legitimate. That’s the whole reason they added that set-aside to RGGI. I think it’s overly complicated and we don’t need to do that for green marketing firms, but RGGI regulators did – your post does not apply to THOSE voluntary green power purchases, right? Folks in RGGI states who buy green power like to think that because of this carveout, their purchase is “for real”…is that the case?

    • Michael Gillenwater says:


      Great question. First, let me point you to Annex A of one of my 2008 papers in Energy Policy on RECs. You can see an open access version of the article here:

      To summarize, though, the case with the RGGI set aside (or any set aside for voluntary green power market) is a red herring. It says nothing about the voluntary green power market or its impact on emissions. Mostly likely the cap will just be inflated to “stock” the set aside. But the more important point is that its not the green power market that will have reduced emissions, it is retiring the allowances. We can say this because we know the voluntary green power market does not reduce emissions. Retiring “hot air” allowances does not change this. It is equivalent to the U.S. Mint printing a bunch of $100 bills and then immediately shredding them. It does not do anything for the economy or money supply. I could create a billion CO2 allowances in the RGGI allowance tracking registry/database, and then delete them immediately. Doing so does not mean that now I have somehow magically reduced emissions by a billion tonnes.

  2. Morgan Jones says:


    What is your view of this in other countries? Is it the voluntary status of RECS that is problematic? In the UK, renewable generation is backed by REGOs (Renewable Energy Guarantees of Origin) administered by OFGEM. Is the UK situation a ‘regulatory’ rather than ‘voluntary’ market and therefore more credible?



    • Michael Gillenwater says:


      As I point out in the referenced articles, there is no reason I am aware of to expect voluntary RECs in Europe and elsewhere to not suffer from the same faults as those in the USA.

      It is key, though, to distinguish voluntary from compliance RECs to meet mandatory quotas on Load Serving Entities (i.e., utilities). These are really two totally different tradable environmental instruments because they operate in two totally different supply and demand contexts. It was like I said all emissions allowances were the same, when one set were traded under a proper cap and another in a market without any emissions cap. We can call them both allowances, but practically speaking, they are different. If we started calling pyrite “gold” it would not mean that pyrite all the sudden would be worth $1000/oz.

      • Morgan Jones says:

        Thanks for your reply.

        I am still interested in your view of the UK energy market? All renewable supply is backed by government issued guarantees of origin (REGOs). I am not sure whether that would be termed voluntary according to your terminology.

  3. John Cowan says:

    Your mission to expose voluntary RECs is long standing, and commendable. But the title of this piece subjects all green power to your venom when you really aim at voluntary RECs. The thrill and confusion of greenness unfortunately attracts green-washing marketers to sell voluntary RECs. But lets be clear that investments in wind, solar or hydro projects do help the environment with every kWH they generate, and most of us want to encourage that (except the coal industry).

    Since I (like many large and small consumers of electricity) cannot build a green power plant in my back yard or on my roof, I pay a green power producer a premium for my grid mix of kWH. I know he has generated the kWh that match his sales. So I believe I can claim my power is green – because I have a direct relationship with a producer whose carbon free output is audited by third party and who does not separate attributes for separate marketing (I did some such audits of the supplier).

    Please focus future titles so your venom comes down on voluntary RECs. I don’t think you want people to think you (of all people) are against investments in green power. We just need people to understand and select proper market channels for voluntarily supporting such investments.

    • Michael Gillenwater says:


      I am guessing that by the context of your message that it was intended to be addressed to “Michael” not “David”. Although between us we have the two most common names for males in the USA.

      I’ll just respond with two points.

      1) the vast majority of voluntary green power purchase claims are made using voluntary RECs. So they are somewhat the same thing. The other major categories are PPAs (which are not really relevant for residential customers) and green power pricing program (in which you pay a higher tariff for your electricity to claim it is green). Some of these pricing programs are well managed and require that the extra fees go to new investments in RE. But many just take the money and it really has no influence on the amount of RE generated.

      2) I suggest you look at the website with our Scope 2 Open letter for a detailed discussion on how to properly do the GHG accounting for green power claims using a project-based framework. Auditing that simply says that yes, a windmill exists and is producing electricity and that RECs are not being sold twice, does not say anything about its causal relationship to emission reductions. See here:

      Sorry, but just because you pay a premium does not automatically mean that it has any influence on RE investment or generation, which means emissions. The question is what influence does your premium have? Probably somewhere between not much and none, unless you are paying a large premium (e.g., $100/MWh versus $1/MWh), such is the case for some solar programs.

      If you read our studies on this topic, we make it clear in every case that we are highly supportive of RE investments. What we want is more mechanisms that promote effective investments, which are not hampered because why would you bother when you can claim to be green powered for cheap and not worry if it is actually doing any good. Apparently too many companies and REC certification groups are OK with the current status.

  4. James C. says:

    Why is additionality necessary for RECs (e.g., renewable energy projects that address scope 2 emissions) when ultimately the ownership of the stack emissions (e.g., the power plant’s emissions) can never be resolved within a bilateral contract between a renewable energy project developer and consumer of electricity on a shared grid. Additionality is a test applied to a project in order to determine causation of an emission reduction outside of the “system” in question in order to make an emissions claim, but on a shared electricity grid I don’t see how true project based accounting can really deal with the ownership issue of the emissions reduction since most utilities I know are not likely to give up the reduction (e.g., legally) regardless of who thinks they caused it. Ultimately this presents a problem of double counting regarless of who caused the reduction. Thus, Corporates who are investing in any form of REC-based green power should not make global emissions reduction claims simply because the REC instrument does not convey an emissions reduction to its buyer no matter the price or the policy framework through which the project was developed. Many detractors of RECs (you included) seem to want project accounting principles (additionality) to be applied to renewable energy in order to verify causation of emissions reductions, but I can’t see how this will ever be possible. Does transformation of the electricity sector through instruments that allocate consumer demand really need to involve additionality? Can one make an emission reduction claim when the instrument in question does not convey that claim nor does the threshold of additionality provide me ownership of the reduction itself?

Leave a Reply

Your email address will not be published. Required fields are marked *